Mapping of cropland, cropping patterns and crop types by combining optical remote sensing images with decision tree classifier and random forest

Mapping and monitoring the distribution of croplands crop types support policymakers international organizations by reducing risks to food security, notably from climate change and, for that purpose, remote sensing is routinely used. However, identifying specific types, cropland, cropping patterns using space-based observations challenging because different have similarity spectral signatures. This study applied a methodology identify cropland including tobacco, wheat, barley, gram, as well following patterns: wheat-tobacco, wheat-gram, wheat-barley, wheat-maize, which are common in Gujranwala District, Pakistan, region. The consists combining optical images Sentinel-2 Landsat-8 with Machine Learning (ML) methods, namely Decision Tree Classifier (DTC) Random Forest (RF) algorithm. best time-periods differentiating other land cover were identified, then Landsat 8 NDVI-based time-series linked phenological parameters determine over region their temporal indices ML algorithms. was subsequently evaluated images, statistical data 2020 2021, field on patterns. results highlight high level accuracy methodological approach presented together techniques, mapping not only but also when validated at county level. These reveal this has benefits evaluating security adding evidence base studies use countries.